Parameter Constrained Neural Network

Model overview

Parameter constrained neural network classifier using PyTorch with configurable hidden layers to predict taxonomy IDs from product tags. Each taxonomy path is encoded as a p-adic integer and the model minimizes p-adic distance.

5 CV folds

13.58% Mean accuracy

0.1073 Mean F1

0.69227768 Mean p-adic loss

Total train samples: 36,440
Total test samples: 9,110

Key hyperparameters

Prime base(s): 79
Hidden layer sizes: 27
Max tags used: 32

Cross-validation results

Fold	Accuracy	F1	P-adic loss (mean)	Details
0	13.26%	0.1083	0.696639	View details →
1	15.04%	0.1245	0.684872	View details →
2	14.07%	0.1034	0.670267	View details →
3	11.45%	0.0903	0.724559	View details →
4	14.10%	0.1101	0.685051	View details →